Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

AutoSNAP: Automatically Learning Neural Architectures for Instrument Pose Estimation

: Kügler, David; Uecker, Marc; Kuijper, Arjan; Mukhopadhyay, Anirban


Martel, Anne L. (Ed.):
Medical Image Computing and Computer Assisted Intervention - MICCAI 2020. Proceedings. Pt.III : 23rd International Conference, Lima, Peru, October 4-8, 2020
Cham: Springer Nature, 2020 (Lecture Notes in Computer Science 12263)
ISBN: 978-3-030-59715-3 (Print)
ISBN: 978-3-030-59716-0 (Online)
International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) <23, 2020, Online>
Fraunhofer IGD ()
Lead Topic: Individual Health; Research Line: Computer vision (CV); interventional technique; medical application; medical imaging; deep learning

Despite recent successes, the advances in Deep Learning have not yet been fully translated to Computer Assisted Intervention (CAI) problems such as pose estimation of surgical instruments. Currently, neural architectures for classification and segmentation tasks are adopted ignoring significant discrepancies between CAI and these tasks. We propose an automatic framework (AutoSNAP) for instrument pose estimation problems, which discovers and learns architectures for neural networks. We introduce 1) an efficient testing environment for pose estimation, 2) a powerful architecture representation based on novel Symbolic Neural Architecture Patterns (SNAPs), and 3) an optimization of the architecture using an efficient search scheme. Using AutoSNAP, we discover an improved architecture (SNAPNet) which outperforms both the hand-engineered i3PosNet and the state-of-the-art architecture search method DARTS.